Efficient Long-Range Attention Network for Image Super-resolution

03/13/2022
by   Xindong Zhang, et al.
0

Recently, transformer-based methods have demonstrated impressive results in various vision tasks, including image super-resolution (SR), by exploiting the self-attention (SA) for feature extraction. However, the computation of SA in most existing transformer based models is very expensive, while some employed operations may be redundant for the SR task. This limits the range of SA computation and consequently the SR performance. In this work, we propose an efficient long-range attention network (ELAN) for image SR. Specifically, we first employ shift convolution (shift-conv) to effectively extract the image local structural information while maintaining the same level of complexity as 1x1 convolution, then propose a group-wise multi-scale self-attention (GMSA) module, which calculates SA on non-overlapped groups of features using different window sizes to exploit the long-range image dependency. A highly efficient long-range attention block (ELAB) is then built by simply cascading two shift-conv with a GMSA module, which is further accelerated by using a shared attention mechanism. Without bells and whistles, our ELAN follows a fairly simple design by sequentially cascading the ELABs. Extensive experiments demonstrate that ELAN obtains even better results against the transformer-based SR models but with significantly less complexity. The source code can be found at https://github.com/xindongzhang/ELAN.

READ FULL TEXT

page 10

page 11

page 14

page 15

research
11/30/2022

From Coarse to Fine: Hierarchical Pixel Integration for Lightweight Image Super-Resolution

Image super-resolution (SR) serves as a fundamental tool for the process...
research
03/19/2023

Q-RBSA: High-Resolution 3D EBSD Map Generation Using An Efficient Quaternion Transformer Network

Gathering 3D material microstructural information is time-consuming, exp...
research
11/23/2022

GhostNetV2: Enhance Cheap Operation with Long-Range Attention

Light-weight convolutional neural networks (CNNs) are specially designed...
research
01/24/2023

Image Super-Resolution using Efficient Striped Window Transformer

Recently, transformer-based methods have made impressive progress in sin...
research
03/11/2023

Recursive Generalization Transformer for Image Super-Resolution

Transformer architectures have exhibited remarkable performance in image...
research
06/11/2023

TransMRSR: Transformer-based Self-Distilled Generative Prior for Brain MRI Super-Resolution

Magnetic resonance images (MRI) acquired with low through-plane resoluti...
research
08/05/2023

Unfolding Once is Enough: A Deployment-Friendly Transformer Unit for Super-Resolution

Recent years have witnessed a few attempts of vision transformers for si...

Please sign up or login with your details

Forgot password? Click here to reset